run-1/checkpoint-64/trainer_state.json

{
  "best_metric": 0.7326732673267327,
  "best_model_checkpoint": "distilhubert-finetuned-not-a-word2/run-1/checkpoint-16",
  "epoch": 4.0,
  "eval_steps": 500,
  "global_step": 64,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.31,
      "grad_norm": 0.8568354845046997,
      "learning_rate": 2.8711716200450114e-06,
      "loss": 0.6977,
      "step": 5
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.7982807755470276,
      "learning_rate": 5.742343240090023e-06,
      "loss": 0.6961,
      "step": 10
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.32968151569366455,
      "learning_rate": 8.613514860135034e-06,
      "loss": 0.6801,
      "step": 15
    },
    {
      "epoch": 1.0,
      "eval_f1": 0.7326732673267327,
      "eval_loss": 0.6840133666992188,
      "eval_runtime": 1.3445,
      "eval_samples_per_second": 47.602,
      "eval_steps_per_second": 5.95,
      "step": 16
    },
    {
      "epoch": 1.25,
      "grad_norm": 1.6010539531707764,
      "learning_rate": 8.932533929028925e-06,
      "loss": 0.6633,
      "step": 20
    },
    {
      "epoch": 1.56,
      "grad_norm": 0.5679539442062378,
      "learning_rate": 8.613514860135034e-06,
      "loss": 0.6444,
      "step": 25
    },
    {
      "epoch": 1.88,
      "grad_norm": 0.4244280755519867,
      "learning_rate": 8.294495791241145e-06,
      "loss": 0.6419,
      "step": 30
    },
    {
      "epoch": 2.0,
      "eval_f1": 0.7326732673267327,
      "eval_loss": 0.6817398071289062,
      "eval_runtime": 1.3677,
      "eval_samples_per_second": 46.795,
      "eval_steps_per_second": 5.849,
      "step": 32
    },
    {
      "epoch": 2.19,
      "grad_norm": 1.0901663303375244,
      "learning_rate": 7.975476722347254e-06,
      "loss": 0.6205,
      "step": 35
    },
    {
      "epoch": 2.5,
      "grad_norm": 0.7114349603652954,
      "learning_rate": 7.656457653453365e-06,
      "loss": 0.6425,
      "step": 40
    },
    {
      "epoch": 2.81,
      "grad_norm": 1.0210336446762085,
      "learning_rate": 7.337438584559474e-06,
      "loss": 0.5994,
      "step": 45
    },
    {
      "epoch": 3.0,
      "eval_f1": 0.7326732673267327,
      "eval_loss": 0.6920928955078125,
      "eval_runtime": 1.3659,
      "eval_samples_per_second": 46.855,
      "eval_steps_per_second": 5.857,
      "step": 48
    },
    {
      "epoch": 3.12,
      "grad_norm": 0.5598271489143372,
      "learning_rate": 7.018419515665583e-06,
      "loss": 0.5898,
      "step": 50
    },
    {
      "epoch": 3.44,
      "grad_norm": 0.8800140023231506,
      "learning_rate": 6.6994004467716935e-06,
      "loss": 0.5947,
      "step": 55
    },
    {
      "epoch": 3.75,
      "grad_norm": 0.5016202330589294,
      "learning_rate": 6.3803813778778036e-06,
      "loss": 0.599,
      "step": 60
    },
    {
      "epoch": 4.0,
      "eval_f1": 0.7326732673267327,
      "eval_loss": 0.7060089111328125,
      "eval_runtime": 1.3676,
      "eval_samples_per_second": 46.796,
      "eval_steps_per_second": 5.849,
      "step": 64
    }
  ],
  "logging_steps": 5,
  "max_steps": 160,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 3370472201643696.0,
  "train_batch_size": 12,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 9.187749184144037e-06,
    "per_device_train_batch_size": 12
  }
}